Most antibodies do their work in the bloodstream. But others may
be powerless to knock out their disease-causing foes unless the
confrontation takes place inside an intestinal cell, researchers
The unexpected discovery suggests a novel approach for designing
antibody-stimulating vaccines against viruses that infect through
the gut, said Dr. Harry Greenberg, a medical investigator at the
Veterans Affairs Palo Alto Health Care System and chief of the
division of gastroenterology at Stanford University School of
The finding also explains the surprisingly broad effectiveness
of a rotavirus vaccine invented by Greenberg and colleagues when
he worked at the National Institute of Allergy and Infectious
Greenberg and three Stanford colleagues describe the research
in the April 5, 1996, issue of Science. A "Perspectives"
article explaining the significance of this work appears in the
The Stanford researchers made the finding during a study of immunity
to rotavirus, the major cause of childhood diarrhea. Greenberg
is associate chair of Stanford's Department of Medicine and a
professor of gastroenterology.
A Costly Malady
Diarrhea due to rotavirus kills between 800,000 and 1 million
children annually, mainly in Third World countries. In the United
States, this virus causes few deaths but costs over $1 billion
each year as a result of health-care expenses and days missed
from work, Greenberg said.
"When you have something as damaging as this, you have a
basic reason to want to understand immunity to the virus. You
hope that understanding immunity will better equip you to produce
a successful vaccine," Greenberg said.
Many previous studies of rotavirus immunity were carried out in
vitro in artificial settings such as petri plates, Greenberg said.
His group set out to learn how rotavirus immunity works in living
In one experiment, Greenberg and colleagues fed infectious rotavirus
to mice they had treated to produce large quantities of antibodies
to rotavirus proteins. They attached an antibody-producing tumor
to each mouse's back to mimic the natural production of immunoglobulin
A (IgA) antibodies.
Unlike IgG antibodies--the type usually stimulated by vaccines--IgA
can attach to the special epithelial cells lining the intestines,
enter them, and then pass into the gut. These epithelial cells
make up a mucous membrane.
In the experiment, each tumor was designed to produce antibodies
against a single rotavirus protein.
The researchers gauged how well the antibodies worked by measuring
the amount of virus in the animals' feces: A little virus indicated
that the antibody killed the virus before it could reproduce to
a great extent; a lot meant that the antibody failed to stop the
The researchers were surprised to find that antibodies to one
protein, called VP6, were the only ones that thwarted the virus.
This was unexpected because in the earlier in vitro studies, antibodies
to VP6 had never been successful.
Also surprising was the finding that the antibody to protein VP4,
which had stopped the virus in the in vitro studies, failed to
knock out the infection in the mice.
"What we're saying is, in this model, unexpectedly, antibodies
that had no activity in vitro were highly protective in vivo,"
Next, the researchers tried placing the VP6 antibodies directly
in the gut. This time the antibodies had no effect.
What seems to be going on here, said Greenberg, is that some antibodies,
such as those against VP6, work only when they meet their target
inside an intestinal cell. Something happens after they enter
the cell that affects their ability to neutralize the virus.
Implications for Vaccine Design
Since most vaccines stimulate only IgA antibodies, which are unable
to pass through epithelial cells, this finding might prompt vaccine
designers to rethink their strategy for fighting the many pathogens--ranging
from the virus that causes the common cold to HIV--that infect
through mucous membranes, Greenberg said.
"If what we've shown is a generally applicable phenomenon,
it means the host immune system might have another way of interrupting
the many pathogens that infect at mucous membranes--a way that
vaccine designers have not really looked at," Greenberg said.
The effect of location on the activity of certain antibodies might
explain the surprisingly broad effectiveness of the rotavirus
vaccine invented by Greenberg and colleagues. Although the experimental
vaccine was expected to protect against one type of rotavirus,
it appears to protect against several, Greenberg said. The vaccine
is likely to be under study by the FDA in the near future for
use in the United States, he said.
What had mystified researchers about the vaccine was that it could
stop viral variants it was not designed to thwart.
"We find that rotavirus comes in different serotypes, or
'flavors.' Some flavors have one type of a protein called VP7,
others have a different type," Greenberg said. The vaccine
worked against viruses carrying various types of VP7, even though
it stimulated antibodies against only one type of VP7.
Now Greenberg suspects that VP6 holds the answer to the paradox.
Although the experimental vaccine stimulates antibodies against
VP6 and VP7, researchers had assumed that VP6 antibodies played
no role in stopping infection because VP6 had been ineffective
in the in vitro studies.
But Greenberg's new study shows that VP6 antibodies can stop infection
in vivo. And unlike VP7, VP6 invades very little from one virus
to the other.
"It may well be the antibodies against VP6 stimulated by
the vaccine that are granting immunity," he said.
Greenberg's coauthors were Stanford postdoctoral fellows John
W. Bums and Majid Siadat-Pajouh and undergraduate Ajit A. Krishnaney.
The research was funded, in part, by the National Institutes of
Health, the World Health Organization, and the Department of Veterans